Matt Heusser’s Blog
Testing at the Edge of Chaos
Be Careful what you wish for …
I haven’t had much time for blogging lately. A few great, interesting, cool things that are swamping me right now -
- Jim Brosseau’s Book, “Taking Ownership”is completely through the greenlight stage and moving to publication. Yaaay!
- I’m speaking at the Grand Rapid’s Java User’s Group on Tuesday the 18th of September. They literally sent out a call for speaker’s last week, and I responded that I had a couple of lightning talks that I think might string together well for ten or fifteen minutes. So they asked me for an abstract for a full talk … it should be interesting.
- I’m coaching a team of 4 and 5 year old children from Allegan AYSO Soccer. GO ALLEAN TEAM FIVE SILLY FROGS!
- I’m serving as a volunteer and a speaker for GLSEC this year.
More about GLSEC next post, but for the time being, let’s just say I’ve been busy …
Something old …
I just got back from the wedding of a collegue.
In that spirit, I am reminded of the talk I gave last year at the Indiana Quality Conference – something James Bach recently refered to as a “Kick-Ass Podcast.” (No really, his words, not mine.)
It’s at the very bottom of the stack for Creative Chaos, so I thought I would let it bubble up -
The title is “So You’re Doomed” – Here’s a link to the PowerPoint (5MB)
And the Audio (45MB).
The audio is forty-five-ish minutes. I look forward to your feedback!
Conferences – V
I promised to detail my strategy to start attending conferences.
Actually, it’s pretty simple.
1) Get involved in a local user’s group. Meet colleagues, discuss ideas, help other people solve their problems, give presentations and/or join the leadership team. If you don’t have a local user’s group, found one. (No, it is not that hard – I founded the Grand Rapid Perl Mongers in 1998, a user’s group that is still around. How to do that is a different post.)
Sidebar:
The value in the user’s group is networking, but I disagree with a lot of people about networking. Some think it is about collecting a large rolodex or making acquaintances to ‘get’ a job or contract. Well, that might work; there are specific techniques people use to jump from lead to lead.
Personally, I don’t buy it. Why would anyone want to connect to a leach? To network effectively, you need to get to know other people at least well enough to know what they are interested in. Once you do that, you listen to the problems they have and see if you can help. Eventually, people in your network will look forward to calls from you because they know you will be offering to help them in some way – or at least have something interesting to discuss. Perhaps, some day, a sense of reciprocity kicks in, and they mention an opportunity for you. Perhaps not. The point is, now a lot of people know you and your reputation grows.
—> And you’ll use that sidebar reputation in step two.
2) Put your local reputation to good use:
You could …
2a) Look for a regional conference or a national conference in your area.
If you live in a big city like Chicago, Indianapolis, Denver, or Portland you are sure to find a regional conference staffed by volunteers. This conference is probably close enough that there will be no travel costs and no hotel costs. It will be short enough that you can miss a day or two – a max of three days – not a week. And you can get in free in two different
ways; you can either speak or volunteer to help run the conference.
For that matter, there are a lot of big national conferences that are in the same place every year, like the Twin Cities, Minnesota area, the Orlando Florida Area, the Boston Area, SanFranciso , Orange County California, and so on. Again, you can get in by either speaking or, in many cases, even the big conferences allow a limited number of volunteer slots. Also, the big conferences may be as long as a week, but you can usually sign up for “Just the conference days” or “Just the tutorial days” and only miss three or four days of work.
If you live in the United States or Canada and can’t find a local conference, drop a comment on my blog and I will see what I can do.
2b) Eventually, someone in your user’s group will know someone else who is hiring. And if you are an active member of the group, they will have some ability to evaluate you, and possibly recommend you. So here’s the secret: Go to a company that regularly sends it’s technology employees to training. Really, some do exist, especially companies with a rule like “One week
of training every two years.”
In my experience, companies that have some training offered to employees are either better places to work or at least tend to pay better.
Plus, by knowing the people that work for the company, you’ll know a lot more going on. You’ll know the technologies they use, the problems they have, and you’ll know if the employees there like it or hate it. In short, you will know if you want the job or not.
2c) Found your own regional conference. Again, I’ve done this, it’s not as hard as it sounds, but this one actually does take some work. If you check out my comments from around November of 2006, you’ll see some the results of our after action review from our first year of GLSEC.
(Option 2D is to attend a free peer conference, such as WOPR, LAWST, or IWST, but I suspect again that is an entirely different post.)
3) By the end of step two, you are attending a conference, probably every year. The next step is to move up from a regional conference to a national one. A few ways to do that …
3a) When you move to the new job, make training a part of the hiring negotiation – at least every-other year. Ask about the training budget. Be specific. Be firm. Don’t jump to a company that has recently made commitments and then suddenly had a hiring freeze.
3b) Find a national conference that you can drive to and get accepted as a speaker (or volunteer if possible). Then your company only has to kick in the fee for the hotel room.
3c) Based on your ever-expanding network, find a third job and … you get the picture.
3d) Start writing or blogging enough to boot-strap demand. Then ask the conference company to cover your travel expenses. (If you speak for two or more hours, they are much more likely to do this. Speak for a half day and you’ll get some kind of support, anyway.)
To sum up:
If you make attending conferences a goal in the same way that some people make “increasing salary” a goal, you’re pretty sure to get it. If you really want salary, attending conferences is a decent way to get it – you can build connections, reputation, and companies that send people to conferences tend to have more money to spend.
There are possible exceptions.
You could have a niche role small enough that you can’t find local conferences. (data modeling or business intelligence, for example) You could live in a “one company town” in the deep Midwest or mountain area. Outside of the US, you could live in a large town with very little in the way of IT jobs, or you might have to bootstrap a development or testing community.
It’s late. I’m starting to ramble. Anyway, that’s the strategy.
Next time: How to generate ideas …
Tutorials and Training
In a response to my blog post on training, Mark wrote:
I would worry more about the thrashing effect of a developer testing course which did not focus on a specific developer tool set. Choosing a specific tool set allows the teacher to remove many variables from the discussion so they can rapidly teach concepts like “red, green, refactor” without spending time deciding if pyunit, nunit, junit, TestNG, or utplsql is the right choice for the user.
I agree, and here’s what I’m thinking -
A one-day course on “What works in software testing: Or How To be Lazy Without Really Trying” (With Credit Mike Schwern, of course)
Before the course, attendees email a list of the core technologies they work on. Then, we customize the afternoon, perhaps splitting into small groups.
Essentially, the course is “Homebrew Test Automation”++ – as a workshop.
Now, the whole idea isn’t fully baked, but developing ideas with people on the cutting edge is part of what I use this blog for. The risks in the format are medium-sized; people REALLY like stable, repeatable training and start to weird out when I customize training on the fly. Such custom training is also hard to get right and easy to get wrong.
Still, that’s what I’m thinking about right now. What do you think?
Lightning Talks Wrap-Up
Just a quick after-action review from STAREast –
Overall it was a great experience. Alan Page has his lightning talk up on-line in English Prose; Michael Bolton has his as a powerpoint (look for his May 27 Post), and I integrated mine into the post Conferences II.
One thing I tried this year that was different was outlawing power point; each speaker had to present using only words, markers and paper.
Of course, the audience would expect powerpoint, so for each speaker we had an intro slide on a black background with white letters – name, title, company, and logo. Then after that I had a completely black slide.
Switching to completely black messed with the projector; it thought there was no content. So I had to change the slides to very-very-very dark grey.
Overall, I think it was distracting. Next time, I will either allow powerpoint from the presenters, or have none at all. It was a nice experiment, but the middle road is often not the best one. 
Another thing I tried was to shorten then intro/conclusion, run the talks as fast as possible, and, time permitting, shim in a tenth talk. This gave the audience the maximum amount of content for the time.
Sadly, this means we missed a few things (like “Can we get a round of applause for all the speakers?”) and I think the audience missed some closure.
So, as great as Michael Bolton’s second speech was, I think that in the future I will stick with nine.
Oh, and I give away a noisemaker, which never seems to go well. I need to just go buy a gong.
(Some people who were at STAREast will reply “Matt, what are you talking about? The LT’s were great!” And perhaps you are right. Still, I’ve heard a title for people who aren’t self-critical and constantly trying to improve – and that title is ‘Amateur’ …)
Lightning Talk Mojo – I
Breathing readers of Creative Chaos know that I am facilitating Lightning TalksatSTAREast 2007 – astute ones realize that I am a genuine fan of the concept.
I would like to tell you why.
This year, I’ve been doing a good bit of back and forth with the lightning talk speakers – encouraging them to turn off powerpoint, and really talk to the audience.
One speaker wrote back that doing a whole five minute speech without powerpoint assistance would be hard.
And he’s right.
But that’s bad.
Powerpoint is a crutch.
Here’s the deep, dark secret of presentations:
Outside of the basic, introductory, everything-I-say-is-new type of talk, in a typical presentation, your audience will leave with just a few nuggets. Sometimes, you only have one nugget, and spend the entire hour beating the audience over the head with it.
Still, most of the time, the audience is listening for insight. The more insights you can sprinkle into your talk, the better.
So, let’s assume that a good nugget takes about five minutes to explain. That means it should be possible to do ten nuggets in a talk. Of course,different people need different things, so we’ll assume that for any given audience member, only have of the “nuggets” will connect.
Now let’s examine the typical 1-hour auto-content-generated slideware talk for a moment:
Start with time for ten nuggets
Intro – 10 minutes – Subtract two. Eight left.
Conclusion – 10 minute – Subtract two. Six left.
Q&A – 15 minutes – Subtract three. THREE NUGGETS LEFT.
Divide by two, because only half of the ideas will be relevant to any audience member.
That means for a one-hour talk, you get to make have about one-point-five actual insights that can change behavior.
That is a lot of sitting around, waiting for something to happen, and not much happening.
How can we do better?
The Intro/Body/Conclusion/QA style is, well, redundant. If you study Toastmasters, they openly admit this – the whole point is to tell ‘em three times, so your single point comes across.
That might work when you are briefing the boss, but at a technical conference, why prove one point when you could prove ten? Even if half your stuff doesn’t apply, and the audience hates half that does apply, heck – you still get to make two and a half points an hour.
I see at least two problems with this:
First, learning to make a point succinctly in five minutes is hard.
Second, the very cognitive format of powerpoint, with it’s bullets and lists, tends to turn your stuff into marketing-ware. Trying to make one point with powerpoint in one slide (or two) is, well … hard.
My suggestions are -
1) Pschologists have discovered a method called “chunking” that people use to memorize extremely large pieces of material. Essentially, you take a big piece of data and split it into many small groups. For example, if you meet someone who has memorized PI to 1,000 digits, you will find that he probably hasn’t memorized a thousand numbers at all. Instead, he has memorized a hundred-odd “chunks”, where each chunk is five to ten numbers.
So is a one-hour talk a collection of ten “chunks”
2) Look into other communications options like handouts, discussion, or writing on an Easel. Read The Cognitive Value of PowerPoint by Tufte and The Gettysburg PowerPoint Presentation by Norvig.
3) If you really want to use powerpoint, consider the “one big slide per point” approach. That means reading, listening, and watching people who use this technique effectively – Tim Lister is a good one to follow.
4) Get really good at making a single point, making it well, and moving on. A good way to do that is to give lightning talks at conferences …
Have I sold you on lightning talks yet?
If yes, your next question is probably “ok, so how do I get really good at chunking my talk?
More to come.
Post-Script: Paul Graham has an article on similar themes about writing essays – you can find it here. Even if you don’t agree with the guy, you’ll enjoy the read, I promise.)
Upcoming Workshop
I will be running a small private workshop next month – about public speaking – for new speakers.
That’s right; I am giving a talk on how to give a talk.
If you are interested in a pilot or peer review, shoot me an email.
Indy Trip Report
Last month I made it out to Indianapolis for a meeting of the Indianapolis QA Association and a short course on software requirements. Sadly, I tried to make the event a working vacation, which meant I had not quite enough time to spend with technologists, yet also not enough time to really spend with my family.
A couple highlights -
1) The slides from my talk on process improvement are available online. If you open the document in powerpoint, you will see detailed (if not proof-read) lecture notes in the notes window.
2) I got the chance to meet some interesting people from Liberty Mutual, including David Christiansen. Apparently, David liked my talk enough to quote me; that’s always a good sign. David also runs the blog site The Tech Darkside, which has some good material on it.
All told, it was a good trip. The requirements class was a bit too big for one facilitator (33 people), and this slowed down the general pace. In the future, I will either hold the line on attendance or bring in a second facilitator.
Way Over Book-ed …
I know, the past few months haven’t provided the typical Creative Chaos output.
So, here’s what I have been doing -
1) Evaluating Submissions for Lightning Talks at STAREast this year – which I hope to announce this week.
2) Preparing for a private workshop in Orlando, in May,
3) Working on a pre-publication book review for Addison-Wesley,
4) Trying to get around to doing the review of Brian Marick’s Everyday Scripting with Ruby,
5) Lining up Keynote Speakers and Tutorial Speakers for GLSEC 2007. No, recruiting high-level speakers is not all glory; there is a lot of nagging and contracts involved. Though it is, mostly glory …
6) Working on lining up sponsors for GLSEC 2007. This mostly consists of asking polietly, with the ocassional grovelling. I am sad to say, this one is not glory at all. And no, I don’t make a dime on GLSEC, it, plus the local perl user’s group, are my big volunteer work for the industry.
7) Trying to do any writing at all. I think the testing challenge of a few weeks back could be published, if only I would ever get off my duff.
www.xndev.com will lose it’s web host next week. Find a new one.
9) In my free time, I need to get started on my 2008 speaking schedule
Then there’s the kids, family, Church, knights of columbus, and the real work that lets me do this fun stuff. Oh, and something about colored eggs or a bunny or something.
It’s been a whirl.
So, for this month, I have scaled back Creative Chaos, and, yes, this week I am scaling back my regular exercise routine to get some things knocked off that list.
Next on my to-do list: Learn to evaluate opportunities in light of what I can actually accomplish …
Sigh.
One thing you can say about being overbooked: It ain’t boring …
Conference Presentation Idea …
Secrets of Enterprise Development
What can we learn about Project Scheduling from Mr. Scott? Security from Mr. Worf? Architecture from Data, Engineering from Geordi, Analysis from Spock, and Leadership from Kirk? Borrowing from human psycology and using Star Trek as a backdrop, Matt Heusser will cover the mythology of software development, and what happens when you involve real people, with all the faults and foibles they actually have.
No, seriously, this might make a decent talk. Gene Rodenberry, the creator of Star Trek, often admitted that the show was half social commentary. The other shows at the time, like GunSmoke and Bonanza, were historical. In those other shows, you couldn’t show a female as a cowboy or an asian as a business owner – nor could you show the dark side of racism. But in Star Trek, to show racism, you could make the other race an alien, or use a racially diverse cast to show how essentially human and similar we all are when compared to the pandorian wampus-beast.
Also, I think the term “Enterprise Development” is vaguely lame. It usually means “Software Development in a big, dumb, slow company that is producing software a cost center, not an investment or profit center.” So, in the talk, I could address some of the weaknesses of “enterprisy” development with humor, which is about the only way to do it.
The big problem is that this isn’t a conference talk – it’s a lightning talk. It’s five minutes of real material, and then some serious fluffage. So I need real examples from Trek beyond Scotty and the Kobayashi Maru.
So, I have a few, but I don’t want to spoil them. What are your ideas?
